landscape evolution ‘Landscape evolution’ is the term used to describe the ways that landscapes evolve or change over time. Some of the earliest geological insights arose from the realization that the Earth has a history and that the landforms on the Earth's surface evolve over time through the action of weathering and erosion. In the late nineteenth century, when the insights and controversy surrounding Darwin's evolutionary thought were receiving wide attention, the American geomorphologist William Morris Davis proposed a theory of sequential (evolutionary) landscape development which, with other subsequently developed theories of landscape evolution, became the basis of geomorphological research for the next 60 or so years. By the 1960s, however, this type of research (which came to be called denudation chronology, meaning an investigation of the rates and sequences of landscape evolution or denudation) had fallen from favour and been replaced by more detailed studies of processes and small-scale landform evolution. There has since been a return to long-term landscape evolution studies with the advent of new theoretical frameworks, such as plate tectonics, and the development of new analytical, modelling and dating techniques, such as radiometric dating, fission-track analysis, and cosmogenic surface-exposure dating.

Theories of cyclical landscape evolution

Davis's cycle of erosion (also known as the geographical cycle) is a theoretical construct which proposes that humid temperate landscapes are a function of what Davis called ‘the trio of geographic controls’, namely: the geological structure; the geomorphological processes of weathering and erosion; and the stage of landform evolution (essentially equivalent to the passage of time). In the absence of complicating factors, such as tectonic activity, landscapes evolve through a fixed sequence of forms that reflect the interaction of the three controlling factors. Davis believed that the most important determinant of landscape form is stage (as marked by the passage of time) with the implication that the age of the landscape can supposedly be ‘read’ from its form; landforms are therefore said to be ‘time-dependent’.

The developmental sequence in the Davisian scheme is initiated by rapid uplift, giving the river systems sufficient potential energy to erode the landscape which, given tectonic stability, then passes through the so-called youthful and mature forms to old age (Fig. la). Slope angles decline through the cycle, until a peneplain, a low-relief plain close to river base-level, has developed. Another uplift event starts a new cycle and the landscape again passes through the fixed evolutionary sequence of landforms as time passes.

The geographical cycle had widespread following among English-speaking geomorphologists, who came to use the terms ‘youthful’, ‘mature’, and ‘old age’ as descriptors of landform types with all the evolutionary and age connotations that these terms carry. These terms are still occasionally used. There was less support for the Davisian scheme in Europe, where geomorphologists tended to follow Albrecht Penck's landscape evolution scheme. Penck's scheme was also cyclical but did not assume a lack of tectonic activity after the initiation of the cycle and could therefore accommodate ongoing tectonism throughout the cycle (Fig. 1b). It also argued that the cycle is characterized by parallel retreat of slopes, rather than slope decline, the end point of the cycle being the pediplain, a low-relief plain close to base-level. The parallel retreat of slopes through time in the Penck scheme means that the age of the pediplain is time-transgressive or diachronous, that is, the age varies across the surface of the pediplain.

Another theory that, until recently, has received relatively little attention among English speakers is the etchplanation theory of the German Büdel, who theorized that land-scapes evolve, especially in the stable cratonic areas of the humid tropics, largely by weathering (hence, ‘etching’) and removal of the weathering material by sheet wash. The thick weathering mantles that seem to characterize many of the tectonically stable humid tropical areas are evidence of this etchplanation. The resulting low-relief plain is called an etchplain.

Lester King's scheme, formulated initially to understand landscape evolution in southern Africa, can be thought of as a blending of the Davis and Penck schemes. King also argued that a short uplift event triggers a new cycle of landscape evolution in which the landscape passes through a sequence of youthful, mature, and old age forms. Unlike Davis, however, he argued (largely on the basis of the landforms characteristic of semi-arid southern Africa) that the landscape evolves by parallel retreat of slopes, culminating in a pediment if the cycle can run its full course (Fig. 1c).

Dissatisfaction with the various cyclical theories

The dissatisfaction in the 1950s and 1960s that led geomorphological research to shift away from the denudation chronology approach to landscape evolution arose for several reasons. At the most general level, a general dissatisfaction with the scientific methodology of landscape evolution research developed, particularly in relation to how the various theories, and the denudation chronologies derived from them, could be tested. How, for example, could Davis's notion that the youthful–mature–old age sequence, through which a landscape must evolve if tectonic activity did not intervene, be tested? Indeed, was the whole sequence itself testable? It was also felt that Davis's (and King's) simplifying assumption of initial rapid uplift of the land followed by a long period of tectonic stability during which the characteristic sequence of landforms could develop was unrealistic. At the very least, passive isostatic uplift in response to denudational unloading would be a necessary consequence of the erosional development of a landscape (just as uplift follows the unloading of a landscape when an ice sheet melts—the so-called postglacial rebound).

These and other dissatisfactions, plus the emergence in the research community of a strong interest in quantitative geomorphology (perhaps partly spawned after the Second World War by a conviction of the great scientific advances that could be made by a quantitative approach to research) led to the virtual abandonment of the cyclical approach to landscape evolution and a switch to detailed quantitative investigations of form and process in small drainage basins.

In the 1970s, John Hack, an American geologist interested in long-term landscape evolution in the Appalachians in the eastern USA, formulated a non-cyclical theory of landscape evolution that drew heavily on the ideas of Davis's contemporary, Grove Karl Gilbert. In the late nineteenth and early twentieth centuries, Gilbert had undertaken quantitative and semi-quantitative geomorphological research, emphasizing in his interpretations the notion of dynamic equilibrium. Gilbert's approach and findings were somewhat overshadowed by Davis's, partly because of the attractiveness of the Davisian scheme in integrating into one scheme the evolution of the whole landscape, and partly because of the forcefulness and persuasiveness of Davis's teaching and lecturing style. Hack revived Gilbert's notions to propose a model of long-term landscape evolution in which landforms attain a dynamic equilibrium form that balances erosional energy and lithological resistance. Hack proposed that, once attained, this equilibrium landscape form should persist unchanged in the absence of tectonic activity sufficient to disrupt the equilibrium. The landscape in Hack's scheme does not pass through an evolutionary sequence over time but attains an equilibrium form and relief that persist unchanged until the valley bottoms in the uniformly downwasting landscape reach base-level and relief starts to decline progressively. Until this time, the landforms are time-independent.

A contrasting framework for the consideration of landscape evolution has been provided by Harold Crickmay's hypothesis of ‘unequal activity’. Crickmay observed that modern landscapes usually contain fragments of ancient landscape such as ancient river deposits or ancient lava flows that have not been eroded away: that is, erosion is not acting equally across the landscape in a Hack-type dynamic equilibrium. Thus, the rate of river incision over time may be more rapid than the rate of lowering of divides or interfluves. This unequal activity results in an increase in catchment relief over time, a situation that has been widely observed, particularly in places of low tectonic activity.

Other insights and techniques in landscape evolution studies

A new framework for landscape evolution studies has been provided by plate-tectonic theory, which links lithospheric and asthenospheric processes acting over tens of millions of years and the development and evolution of the Earth's topography by tectonic processes, such as uplift. Most importantly, the tectonic framework, which early workers such as Davis, Penck, and King found to some degree problematical in their interpretations, has been provided by plate tectonics, independently of the landscape evolution theories themselves. The long-term evolution of river systems can be linked to plate-tectonic processes and tied in turn to the development of slopes and interfluves. This has been done successfully for both passive continental margin areas, such as southern Africa, south-eastern Australia, and eastern North America, and tectonically active plate-collision areas, such as the Southern Alps in New Zealand. Areas distant from plate margins, such as the continental cratonic interiors, have been less amenable to such analysis, reflecting the tectonic stability of the cratons and the strong emphasis in plate tectonic theory on tectonically driven changes at the Earth's surface. Cratonic regions have generally continued to be interpreted in cyclical terms, most recently in the context of Büdel's etchplanation theory.

A range of analytical techniques has made it possible to date land surfaces for landscape evolution studies. Data on rates of long-term lowering of the landscape are provided by fission-track analysis, which determines the time when a sample found at the ground's surface passed through a certain temperature (equivalent to a depth) in the underlying crust (Fig. 2a), and by empirical relationships between the depth of mineral formation and the age of the mineral. Palaeomagnetic and radiometric dating of ancient materials (such as lavas and weathering profiles; Fig. 2b) has made it possible to assign absolute ages to land surfaces. In general this has indicated that landforms are older than had been previously thought using the classic cyclical approaches to landscape evolution. The greater-than-anticipated ages mean that rates of denudation of the land surfaces since these dated materials were deposited or formed are lower than had been previously thought, especially in areas of low tectonic activity or those that did not experience Quaternary glaciation.

Powerful computer-based numerical models have been developed that link lithospheric and asthenospheric tectonic processes with surficial processes of weathering and erosion (generally in a plate-tectonic context). These models have provided significant insights into long-term landscape evolution. They have been able to simulate the topography observed along passive continental margins, for example, highlighting the relative importance of tectonic processes and surficial processes (weathering and erosion) in generating topography. They have also shown how denudational and flexural isostatic processes modify topography. These models are being used to elucidate landscape evolution at the temporal and spatial scales considered by Davis, Penck, and King, but we are not yet able to integrate all aspects of landscape evolution into a grand theory. At this stage, we probably know more about rates of landscape evolution and the important roles of passive and active tectonics than about the precise forms through which the landscape passes during its evolution.

Paul Bishop

Bibliography

Summerfield, M. A. (1991) Global geomorphology, Chapter 18. Longman, Harlow.

Landscape evolution

A landscape is the cumulative product of interaction among dynamic geological processes over time. A region's topography and suite of characteristic landforms are, thus, clues to its geologic history. For example, the landscape of rugged, linear mountain chains , deep canyons, dry lake beds, and mesas in the United States'desert southwest tells a geologic story of fluvial and Eolian erosion acting during a period of increasing climatic aridity while plate tectonic forces caused crustal extension and uplift. Earth processes carve a landscape; dynamic interactions between processes control its evolution over time.

The earth's internal heat drives plate tectonic motion and influences the related processes of crustal uplift, magmatic intrusion, volcanism, crustal deformation, and seismic activity. External heat from the Sun forces circulation of Earth's atmosphere and hydrosphere, which in turn drives sedimentary processes such as weathering, erosion, transportation, and deposition. These forces, interacting under the influence of gravity , shape Earth's surface.

Earth processes interact in complex feedback systems. A change in the rate or directional alignment of one process—for example, an increase in rainfall or the abandonment of a river channel—may start a cascade of compensatory changes throughout a region. Plate-tectonic mountain-building and erosion interact in a negative feedback system that regulates the elevation of continental mountain belts. Elevation interacts with temperature and rainfall, the components of climate , to regulate rates of erosion. Climate interacts with vegetation to create soils. A balance between precipitation and temperature maintains a glacier. These are just a few examples of the dynamic processes that shape a regional landscape, and of the interactions that remold an existing array of landforms over time.

See also Eolian processes; Weathering and weathering series